See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/327337318

Customer Churn Analysis and Prediction in Telecommunication for Decision

Making

Conference Paper · August 2018

CITATIONS READS

2 2,051

3 authors, including:

Banage T. G. S. Kumara
Sabaragamuwa University of Sri Lanka
43 PUBLICATIONS   146 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Service clustering and Applications View project

All content following this page was uploaded by Banage T. G. S. Kumara on 01 March 2019.

The user has requested enhancement of the downloaded file.

 2018 International Conference On Business Innovation (ICOBI), 25-26 August 2018, NSBM, Colombo, Sri Lanka
Customer Churn Analysis and Prediction in
Telecommunication for Decision Making
P.K.D.N.M. Alwis B.T.G.S. Kumara
Department of Computing and Department of Computing and
Information Systems Information Systems
Sabaragamuwa University of Sri Sabaragamuwa University of Sri
Lanka Lanka
Belihuloya, Sri Lanka, Belihuloya, Sri Lanka
madushani.niroshi@gmail
Abstract— With the rapid development of communication
technology, the field of telecommunication faces complex
challenges due to the number of vibrant competitive service
providers. Customer Churn is the major issue that faces by the
Telecommunication industries in the world. Churn is the activity
of customers leaving the company and discarding the services
offered by it, due to the dissatisfaction with the services. The
main areas of this research contend with the ability to identify
potential churn customers, cluster customers with similar
consumption behavior and mine the relevant patterns embedded
in the collected data. The primary data collected from customers
were used to create a predictive churn model that obtain
customer churn rate of five telecommunication companies. For
model building, classified the relevant variables with the use of
the Pearson chi-square test, cluster analysis, and association rule
mining. Using the Weka, the cluster results produced the
involvement of customers, interest areas and reasons for the
churn decision to enhance marketing and promotional activities.
Using the Rapid miner, the association rule mining with the FP-
Growth component was expressed rules to identify
interestingness patterns and trends in the collected data have a
huge influence on the revenues and growth of the
telecommunication companies. Then, the C5.0 Decision tree
algorithm tree, the Bayesian Network algorithm, the Logistic
Regression algorithm, and the Neural Network algorithms were
developed using the IBM SPSS Modeler 18. Finally, comparative
evaluation is performed to discover the optimal model and test
the model with accurate, consistent and reliable results.
Keywords—bayesian network, c5.0 decision tree, logistic
regression, neural network
I. INTRODUCTION
Decision making is a key feature of every organization.
The quality of decisions made is dependent on some amount
of knowledge generated from existing or researched
information. The use of modern analytical tools to generate
such knowledge is reasonable for any profit-driven firm.
Taking decisions on customers is one of the key points in
most companies, especially companies in the service sector.
The ability of these companies to predict customer churn is
critically inadequate. Customer churn is the action of the
customer who is like to leave the company and it is one of
the mounting issues of today’s rapidly growing and
competitive telecommunication industry. To minimize the
customer churn, prediction activity to be an important part of
the telecommunication industry’s vital decision making and
strategic planning process.
40
H.A.C.S. Hapuarachchi
Department of Computing and
Information Systems
Sabaragamuwa University of Sri
Lanka
Belihuloya, Sri Lanka
A. Churn Prediction
Today numerous telecom companies are prompt all over
the world. Telecommunication market is facing a severe loss
of revenue due to increasing competition among them and
loss of potential customers [1]. Churn is the activity of the
telecommunication industry is the customers leaving the
current company and moving to another telecom company.
Many companies are finding the reasons of losing customers
by measuring customer loyalty to regain the lost customers.
To keep up with the competition and to acquire as many
customers, most operators invest a huge amount of revenue
to expand their business in the beginning [2]. In the
telecommunication industry each company provides the
customers with huge incentives to attract them to switch to
their services, it is one of the reasons that customer churn is a
big problem in the industry nowadays. To prevent this, the
company should know the reasons for which the customer
decides to move on to another telecom company. The
Telecom Churns can be classified into two main categories:
Involuntary and Voluntary. Involuntary are easier to identify.
Involuntary churn is those customers whom the Telecom
industry decides to remove as a subscriber. They are churned
for fraud, non-payment and those who don‘t use the service.
Voluntary churn is difficult to determine because it is the
decision of the customer to unsubscribe from the service
provider. Voluntary churn can further be classified as
incidental and deliberate churn [3]. The former occurs
without any prior planning by the churn but due to change in
the financial condition, location, etc. Most operators are
trying to deal with these types of churns mainly.
B. Churn Management
Churn management is very important for reducing churns
as acquiring a new customer is more expensive than
retaining the existing ones [4]. Churn rate is the
measurement for the number of customers moving out and in
during a specific period of time. If the reason for churning is
known, the providers can then improve their services to
fulfill the needs of the customers. Churns can be reduced by
analyzing the past history of the potential customers
systematically [5]. A large amount of information is
 2018 International Conference On Business Innovation (ICOBI), 25-26 August 2018, NSBM, Colombo, Sri Lanka

maintained by telecom companies for each of their customers The type of customer, whether a
4 Tariffs
prepaid or post-paid customer
that keep on changing rapidly due to a competitive Length of time a customer has
5 Tenure
environment. The information includes the details about been with a particular subscriber
billing, calls and network data. The huge availability of Approximates the amount used to
Credit purchase
6 purchase call credits a month in
information arises the scope of using Data mining techniques amount (CpM)
rupees
in the telecom database. The information available can be Data purchase
Approximates the amount used to
7 purchase data bundles a month in
analyzed in different perspectives to provide various ways to amount (DpM)
rupees
the operators to predict and reduce churning. Only the Identifies whether customer have
8 Internet usage
relevant details are used in the analysis which contributes to used internet facility or not
Determines whether product
the study from the information given. Data mining 9 Product innovation innovation is necessary for
techniques are used for discovering the interesting patterns sustaining customers
within data and it helps to learn to predict whether a Identifies whether customer have
10 Churn
changed networks or not
customer will churn or not based on customer‘s data stored
in the database.
B. Data Pre-processing
C. Research Objectives The training and testing dataset used in this research may
The main objective of this research is to produce a be included missing data, repeated data or inconsistent data.
predictive model with better results that assess customer To handling missing data and removing duplicated data
churn rate of telecommunication companies using the values data pre-processing is done. The RapidMiner tool is
predictive analytics algorithm for data mining. used at this stage to pre-process the data for analysis and
mining. In doing cluster analysis, the Pearson chi-square and
The supporting objectives examined are to: predictive model building, the data types to be converted into
i. Cluster customers into various categories to numerical values.
enhance marketing and promotional activities.
ii. Mine the relevant patterns embedded in the TABLE 2: CODES FOR ALTERNATIVES

collected data have a huge influence on the Variable Alternative Code

revenues and growth of the Telecommunication Female 0
Gender
Male 1
companies. Student 1
Government Employee 2
II. METHODOLOGY AND EXPERIMENTAL Occupation Private Employee 3
DESIGN Own Business 4
Others 5
Data mining and statistical algorithms were used in the Dialog 1
data analysis, model building and model deployment in this Mobitel 2
Network often used Airtel 3
research. Weka 3.8, Minitab 17, RapidMiner Studio 8.1 and Hutch 4
IBM SPSS Modeler 18 were the analytical tools used in the Etisalat 5
respective analysis and mining process. Less than 1 year 1
Tenure 1-3 2
3-5 3
A. Data Collection Above 5 4
No 0
The questionnaire was used as the tool to collect the data Churn
Yes 1
primarily from customers. The Google drive plug-in was Pre-paid 1
used to design the questionnaire. Training data was collected Tariffs Post-paid 2
Both 3
from the 200 respondents and 50 responses were received
No 0
from respondents on the questionnaire for testing data. The Usage of Internet
Yes 1
data was collected during the period (October – November) No 0
of 2017. Product innovation Yes 1
Not sure 2
C. Research Framework
TABLE 1: THE VARIABLES USED IN DATASET FOR THIS RESEARCH In figure 1, a research framework developed to address
No Variable Name Description problems of this research. This research framework details
Age, Gender, the sectorial areas of concentration and the data mining
1 Demographic variables considered
Occupation algorithms adapted in creating the predictive model. It
Identifies the number of mobile
The number of includes a model deployment and evaluation strategies that
2 networks a customer is connected
networks
to and actively using will assess its effectiveness and efficiency.
Frequently used Identifies the most frequently used
3
network mobile network by the consumer

41
 2018 International Conference On Business Innovation (ICOBI), 25-26 August 2018, NSBM, Colombo, Sri Lanka
Figure 1: Research framework of Customer Churn
Analysis and Prediction in Telecommunication
III. RESULTS
A. Pearson Chi-square Test
Pearson Chi-square test is used to evaluate the variables
which are associated with the decision of churn that can be
used in the predictive model building. Pearson and likelihood
ratio chi-square tests are conducted using Minitab. The test
produced significant results (p–value is less than α level of
0.05) to indicate that some of the variables have an
association with the decision to churn.
TABLE 3: SUMMARY OF ASSOCIATION OF EACH ATTRIBUTES THE
CHURN DECISION
A. C
Variable P-Value Association
Marital Status 0.645 No l
Gender 0.038 Yes u
Age 0.005 Yes s
Occupation 0.011 Yes
Monthly Income 0.001 Yes t
Purpose of mobile phone e
0.107 No
usage r
No of mobile network
connected
0.003 Yes A
Mobile network often used 0.006 Yes n
Tenure 0.000 Yes a
CpM 0.004 Yes l
Tariffs 0.029 Yes
y
Internet usage 0.020 Yes
DpM 0.105 No s
Product Innovation 0.021 Yes i
s patterns and trends between variables in the dataset. It is
Cluster Analysis is used to discover groups with identical
features in collected data. These groups explained the interest
areas and churn decision with the reasons for targeted
marketing and product development. Using Weka 3.8 the k-
42
means clustering produced four clusters out of the 200
collected data.
TABLE 4: CLUSTER INSTANCES WITH PERCENTAGE
Cluster Clustered Percentage
Number Instances (%)
Cluster 0 69 17
Cluster 1 34 32.5
Cluster 2 65 16
Cluster 3 32 34.5
TABLE 5: FINAL CLUSTER CENTROIDS
Variabl Full Cluster Cluster Cluster Clust
e Data 0 (69.0) 1 2 er 3
(34.0) (65.0) (32.0
(200.0)
)
Gender 0.515 1 1 0 0
Age 33.24 35.49 54.79 22.93 26.40
Occupati
2.305 2.6232 4.0882 1.18 2
on
Monthly 38833.3 61617. 10984. 2740
31827.5
Income 3 64 61 6.25
NoOfMo
bileNet
2.125 2.5217 3 1.30 2
workCo
nnected
MobileN
etworkO
2.195 2.2609 4.52 1 2
ftenUse
d
Tenure 3.085 3.7391 4 1.95 3
Tariffs 1.325 1.1884 1.38 1.35 1.5
CpM 1003.25 1105.79
1948.5
504.61
790.6
2 2
Internet
0.8 1 1 0.3846 1
Usage
Churn 0.675 1 1 0 1
Telecom providers can leverage this cluster model to
allocate customers’ for conducted promotional activities. It is
observed in the clusters that the churners are mostly
businessman and private employees who are generally males
and government employees who are female. These churners
spend a lot of call credit per month and used prepaid service
package. The customers do not intend to churn are mostly
students who are generally females. Telecom providers,
especially those who have endured a churn of customers
need to pay attention in this cluster to the reason for churn as
presented by these customers.
B. Association Rule Mining
Association rule mining used to determine interestingness
contracted to identify strong rules explored in the dataset
using some measures of interestingness. The RapidMiner
Studio 8.1 tool was used in creating the Association rules
model for collected data. In creating the model, the Frequent
 2018 International Conference On Business Innovation (ICOBI), 25-26 August 2018, NSBM, Colombo, Sri Lanka
Pattern Growth (FP-Growth) algorithm was used to mine
associations between variables that result in a churn decision
with particular interest and focus on confidence. The
generated Association rules model presented in Figure 2.
Tariffs=Prepaid
and
Gender=Female
B. Predictive Model Building
Using the valid variables identified in the Pearson Chi-
square test, the four predictive models are created with IBM
SPSS Modeler 18.0 data mining software. The four
classification modeling techniques; C5.0 tree, the Bayesian
network, Neural Network and Logistic regression are used to
create predictive models. The optimal model is
recommended based on individual models and performance
metrics.

Figure 2: Association rules model
In the Table 6 showed ten (10) generated association rules
were selected based on filtering the conclusion as the
decision of churn is yes and sorted in descending order in
line with confidence. The sorted rules have a maximum
confidence of 95.5 percent and a minimum of 84.6 percent.
An auto classifier was applied in the created C5.0
tree model in Figure 3, to test whether the selected C5.0
algorithm will be determined as one of the best algorithms to
create the predictive model.

TABLE 6: TOP (10) GENERATED RULES

Confidence
Conclusion
Premises

Support

Laplace
No

InternetUsage=
Yes, Gender
Churn_Ye
1 =Male and 0.105 0.955 0.995
Tenure = 3-5
s Figure 3: C5.0 algorithm tree model
years
Tariffs=Prepaid, The C5.0 algorithm was listed in the suggested
Gender=Male Churn_Ye
2 0.100 0.952 0.995 churn algorithms which were applied to the data. In Figure 3,
and Tenure= 3- s
5 years the matrix was applied to create a table showing the
Gender=Male
3 and Tenure=3-5
Churn_Ye
0.130 0.929 0.991
relationship between fields of Churn by $C-Churn. In the
s
years created above model, analysis and evaluation are used to
MobileNetwork create a report and a chart for comparing the accuracy of
OftenUsed=Mo
Churn_Ye predictive models.
4 bitel and 0.115 0.920 0.991
s
Tenure=3-5
years
InternetUsage=
Yes and
Churn_Ye
5 MobileNetwork 0.105 0.913 0.991
s
OftenUsed=Mo
bitel
Tariffs=Prepaid
Churn_Ye
6 and Tenure= 3- 0.230 0.902 0.980
s
5 years
InternetUsage=
Yes, Figure 4: Bayesian network Figure 5: Neural network
Churn_Ye
7 Tariffs=Prepaid 0.190 0.884 0.979 model model
s
and Tenure= 3-
5 years
Tenure= 3-5 Churn_Ye
8 0.270 0.871 0.969
years s
Tariffs=Prepaid,
Gender=Female Churn_Ye
9 0.130 0.867 0.983
and Tenure= 3- s
5 years
1 InternetUsage= Churn_Ye
0.110 0.846 0.982
0 Yes, s
Figure 6: Logistic regression
model

43
 2018 International Conference On Business Innovation (ICOBI), 25-26 August 2018, NSBM, Colombo, Sri Lanka
and effectiveness of the model in predicting customer churn
in Telecommunication.
As the result of generating the logistic regression
model, it built up a statistical model which consists of two no yes % correct
mathematical equations to calculate the ability of a person C5.0 no 45 20 69.2
being churner or non-churner. yes 10 125 92.5
Overall Percentage 85%
no yes % correct
Equation 1: Calculating Y’ BN no 47 18 72.3
yes 24 111 82.2
Y' = 0.0682*Gender+(-0.00182)*Age
Overall Percentage 79%
+0.04558*Occupation+0.00001458* no yes % correct
MonthlyIncome+(-0.7214)*Tenure LR no 27 38 41.5
+(-0.2053)*Tariffs+0.00001024*CpM+2.659 yes 18 117 86.6
Overall Percentage 72%
Equation 2: Calculating P(1) no yes % correct
NN no 21 44 32.3
P(1) = exp(Y')/(1 + exp(Y')) yes 16 119 88.1
Overall Percentage 70%

Equation 1 consists of most relevant variables TABLE 8: ACCURACY AND AUC VALUE OF EACH MODEL
which are most affected by the churn decision. The variables
Contrasting the four models, the C5.0 algorithm of
values should be replaced by this equation and then the value
decision tree proved optimal model with 85% accuracy and
of Y' can be calculated. Then the calculated Y' value should
AUC value as 0.888 for the customer churn analysis and
be replaced with the equation 2 and calculate the value of
prediction in Telecommunication based on the chosen
P(1). Prediction of being a churn or non-churn customer is
variables and attributes.
depending on this P(1) value.

If the P(1) value is equal or greater than 0.5, then the D. Model Testing
prediction result is positive and the person will be a churner. The optimal model based on the results of the evaluation
If the P(1) value is less than 0.5, the result is close to 0 is tested on the dataset designed to test the model. The C5.0
(zero). It means the prediction result is negative and the algorithm model was used to test the data as it was identified
person will be a non-churner. as the most optimal among the models. The chosen optimal
model was tested using the test data collected from
C. Model Evaluation customers. The test data has 50 observations, 7 variables and
The four models are evaluated by testing the significance coded the same as the coding in Table 2. The distribution of
of the predictive model generated. The performance metrics the dataset is along with all the gender, age, monthly income,
of all the models were correlated for optimal performance occupation and the other demographic and operational
using Area Under Receiver Operating Characteristic Curve variables used to develop the model. Predictions are then
(AUROC). made to indicate which customers are likely to churn and
those that are not. The predictor variable and target variables
TABLE 7: CONFUSION MATRIX WITH TRAINING DATA
used in building the predictive churn model were tested for
significance.
Model Accuracy (%) AUC Value
C5.0 algorithm
85 0.888
model
BA model 79 0.886
LR model 72 0.762
NN model 70 0.759

The variables were equally tested for validity and

reliability. The validity of the model indicates that it
measures what it is intended for while reliability test
produces consistent results. The tests assessed the efficiency

44
 2018 International Conference On Business Innovation (ICOBI), 25-26 August 2018, NSBM, Colombo, Sri Lanka
Figure 7: Test model for C5.0 algorithm
The test data is applied by mapping the dataset to the model
designed by the C5.0 algorithm as indicated in Figure 7.
Further model screening and applications are initiated to
define the output in determining the likelihood of churn. The
test results presented as the model predicted that 36
customers will churn with confidence from 100% to 55.6%.
It was further explained by the results that over 62% of the
churn customers have a confidence of above 80%. According
to the Figure 8, the results also indicate the churn customers
staying their network above 5 years. It is expensive to
acquire new customers than to retain existing ones, the
prediction of churners and the reasons proffered earlier need
close attention. The top 10 churners and non-churners
predicted by the model are presented in Figures 8 and 9
respectively. The source of the test data set can be connected
to the database or server of the company to produce a real-
time output of churn results for decision making.
Figure 8: Results of test predictions_Yes
Figure 9: Results of test predictions_No
V. DISCUSSION AND CONCLUSION
Data mining is a symbolic tool in the
Telecommunication industry that can exploit the large
volume of data generated for pattern analysis. The recent
increasing embrace of the predictive algorithm of data
mining has given room for companies to assess their future
success, challenges, and targets. The research brings to fore
the relevant untapped customer data and knowledge for
churn prediction and customer classification for better
decision making. Clustering customers were developed in
this research to determine the involvement of customers,
interest areas and reasons for the churn decision. The results
of the cluster analysis can be used in promotional and direct
45
View publication stats
marketing purpose to access marketing strategies in the
industry. In addition, the association rule mining was
provided the significant results that present relevant
knowledge of factors that have a huge influence on the
revenues and growth of the Telecommunication companies.
Telecommunication companies must grasp on this finding
and work to maintain their clients. C5.0 Decision tree model,
the Bayesian Network model, Logistic Regression model,
and the Neural Network model were used and compared for
the most optimal model that predicts accurately. The C5.0
algorithm of decision trees model proved optimal among the
models with 85 percent accuracy and AUC value as 0.888.
The C5.0 algorithm model of the decision tree can be
recommended for churn management. The models can be
used by industry with the IBM SPSS Modeler or any other
appropriate tool with the same algorithm. The
Telecommunication companies can connect the models
directly to their servers or database to produce real-time
results.
ACKNOWLEDGMENT
This study was done by the Department of Computing
and Information Systems in the faculty of Applied Sciences
in the Sabaragamuwa University of Sri Lanka. The author
would like to thanks to the Dr.B.T.G.S.Kumara (Head,
Department of Computing & Information Systems,
Sabaragamuwa University of Sri Lanka) and
Mr.H.A.C.S.Hapuarachchi (Lecturer, Department of Sports
Sciences and Physical Education, Sabaragamuwa University
of Sri Lanka) for their contribution in stimulating
suggestions and encouragement, assistance, guidance, and
cooperation.
REFERENCES
[1] V. Umayaparvathi and K. Iyakutti, "A Survey on Customer Churn
Prediction in Telecom Industry: Datasets, Methods and Metrics,"
International Research Journal of Engineering and Technology
(IRJET), vol. 03, no. 04, April 2016.
[2] Shin-Yuan Hung and Hsiu-Yu Wang, "Applying Data Mining to
Telecom Churn Management," Department of Information
Management, National Chung-Cheng University, Taiwan, ROC,.
[3] M.Balasubramanian and M.Selvarani, "Churn Prediction in Mobile
Telecom Systems Using Data Mining Techniques," Department Of
Computer Science, Annamalai University, Chidambaram, April 2014.
[4] Rahul J. Jadhav and Usharani T. Pawar, "Churn Prediction in
Telecommunication Using Data Mining Technology," International
Journal of Advanced Computer Science and Applications, vol. 2, no.
2, February 2011.
[5] K.Dahiya and S.Bhatai, "Customer churn analysis in telecom
industry," 4th International Conference on Realibility, Infocom
Tehnilogies and Optimization(ICRITO), 2015.
[6] Amjad Khan and Zahid Ansari, "Comparative Study Of Data Mining
Techniques In Telecommunications-A Survey," Dept of Electronics
and Communication, P.A. College of Engineering, Mangalore, India.